This invention relates to medical device systems such as cardiac pacing systems and, more particularly, implantable systems the having the capability of sensing the occurrence of cardiac signals and distinguishing such signals from noise.
Implantable medical devices such as cardiac pacemakers and defibrillators have a great need to accurately and reliably obtain cardiac signals, both for controlling the response and operation of the device and for collecting diagnostic information. This need results in placing substantial design importance on the ability to process such signals so as to determine when a genuine cardiac signal has been received, and to invalidate signals that represent noise or other artifacts. Thus, for an implantable pacemaker or pacemaker-cardioverter-defibrillator type device, it is important to sense all natural activity, including the wave components of normal conducted rhythms as well as premature ventricular beats (PVCs). But at the same time, it is important to distinguish artifacts, including muscle potentials; polarization; electromagnetic interference; lead artifacts; environmental electrical noise; and low level bio-electrical noise picked up by the system. This requirement of sensing the true signals and rejecting noise applies to each signal channel, e.g., both P waves and R waves for dual chamber pacemakers, as well as other signals in more complex systems.
The basic technique used in the pacemaker art is to establish a signal threshold, or sensitivity level, that is below the expected signal level for the signal being sensed, but above the normal noise level. Take as an example the task of detecting R waves of a normal amplitude range of 2.5 to 5.0 mV, with a signal channel having a noise level in the range of 0.25 to 1.0 mV. In such case, the threshold for a signal may be set somewhere between 1.5 and 2.0 mV, so as to eliminate detection of noise but insure sensing of valid signals. However, a problem is that signal levels will change, as will noise levels, in which case the sensitivity, or signal threshold, must be adapted accordingly. The prior art has shown many designs for doing this, most of which employ some manner of collecting signal and amplitude data and adjusting the threshold to stay above the noise level indicated by the recent data. Another problem that must be dealt with is that of adjusting gain so as to avoid excessive clipping of the signals, and to make the full amplifier range available for signal detection. If there is excessive clipping, then it is not possible to obtain an accurate profile of signal amplitudes. Further, if morphology analysis is to be used in event identification, then it is critical to contain clipping to within certain limits. Accordingly, the need in the art is to adjust gain in order to optimize amplification without too much distortion, and to adjust sensitivity in order to discriminate the true signals from noise.
A major problem that continues to confront design in this area is how to optimize the use of data so as to efficiently and reliably distinguish valid event signals from noise or other artifacts. Some prior art systems vary threshold on the basis of the signal and noise measurements of the last cycle, but such a technique is vulnerable to making adjustments in response to incidental artifacts and noise occurrences. Even systems that accumulate and average data over more than one cycle usually do not collect and process the data in a way that permits optimum profiling of signal amplitudes and noise.
Examples of prior art medical device systems with gain control or sensitivity adjustment schemes are found in the patents listed in Table 1 below. Note that it is not admitted that any of the patents listed in Table 1 necessarily constitute prior art with respect to the present invention.
All patents listed in Table 1 above are hereby incorporated by reference herein in their respective entireties. As those of ordinary skill in the art will appreciate readily upon reading the Summary of the Invention, Detailed Description of the Preferred Embodiments and Claims set forth below, many of the devices and methods disclosed in the patents of Table 1 may be modified advantageously by using the teachings of the present invention.
The present invention has a number of objects relative to the current state of the art. The various embodiments of the present invention provide solutions to one or more problems existing in the area of medical devices and, more particularly, implantable medical device systems and methods for accurately and reliably sensing and identifying patient cardiac signals.
One object of the invention is to provide one or more signal channels for sensing cardiac signals, and for obtaining and storing sample data representative of signal amplitude and noise amplitude over a period of time, the data being sampled and arranged so as to optimize calculation of channel sensitivity to discriminate valid signals from noise. The object is achieved by sampling both signal and noise amplitudes over time periods long enough to avoid distortion of information due to transient noise or artifacts, and storing the data over time periods suitable for accumulating sufficient data from which to accurately determine the current signal amplitudes and noise amplitudes so they can be compared.
Another object of the present invention is to arrange the sampled data in a form that enables efficient comparative analysis of the signal and noise data, and periodic adjustment of the data to reduce the long term historical impact of prior data, thereby continuously providing updated data that reflects true changes in the level of the patient signals without being unduly influenced by recent artifacts or past signal levels. The end objective of the signal and noise data analysis is periodic adjustment of the sensitivity level of the signal channel.
Another object of the present invention is to provide gain adjustments that accompany the sensitivity adjustments, the gain adjustments being based on collected signal data that contains information relating to signal clipping. The gain data and gain adjustments are preferably carried out in a manner that enables accurate adjustment of the sensitivity level to correspond to any change in gain level.
In accord with the above objects, there is provided a system and method of obtaining and processing signal data, adapted for use in one or more channels of a medical device such as a pacemaker or other implantable cardiac device. The system is based on using DSP circuitry that provides filtered and unfiltered wave signal and noise signal parameters, which parameters are stored and processed by a microprocessor system to determine desired adjustments in channel gain and sensitivity setting. The system employs the construction of three histograms for each signal channel. A first histogram is built to store data representative of unfiltered signal amplitudes, which data is used to determine percentage of signal clipping in order to indicate desired gain adjustment. Gain is suitably adjustable to one of a predetermined number of levels, and the gain histogram has a corresponding number of bins, each bin representing unfiltered maximum signal amplitudes within an amplitude range. The histogram bin widths are set so that the percentage of clipped signals can be easily determined, providing a basis for gain adjustment. The histogram bins match the gain settings. Each gain adjustment is accompanied by a shift of the data in the histogram bins so that the resulting bin data corresponds to the new gain.
In another embodiment, a quick gain adjustment feature is provided, in order to provide quicker response to changes in the cardiac signal levels during the acute situation just following implant of the device. This feature can be programmed for a predetermined period, for example, a week or two, and is based simply on adjusting gain whenever a predetermined percentage of clipping has been found over a given time period, or given number of signal events.
A second histogram is continuously maintained to provide signal amplitude data received from the DSP circuitry. The signal amplitude data is taken a programmable sample intervals, e.g., every 6 minutes, designed to be long enough to avoid the influence of bursts of invalid signals. The signal amplitude data is collected in the signal histogram over a period of time set to provide statistically valid data, e.g., every 7 days. Concurrently, noise data is sampled at the same intervals, and stored in a noise histogram. When the 7 day interval is timed out, and sufficient data has been collected, the signal and noise histograms are compared, and an algorithm determines whether an adjustment in channel sensitivity is indicated. If such an adjustment is indicated, it is made, and the corresponding new digital threshold levels are set in the DSP circuitry. The gain adjustment routine is coordinated with the sensitivity adjustment.
The object of optimizing use of signal and noise data is facilitated by the technique of building three histograms for each signal channel of the system. The influence of data over time can be determined by dividing the contents of each histogram by a predetermined factor (multiplying by a factor less than 1) following each new gain and sensitivity analysis. Alternately, past bin data can be adjusted by keeping only counts above a certain level. The histogram data is of a form that it can be efficiently analyzed to avoid basing the sensitivity settings on statistically insignificant samples.